ValueError: shapes (256,33626) and (33227,800) not aligned: 33626 (dim 1) != 33227 (dim 0)

[ouening]

I use the trained model through:

classifier = SupervisedDBNClassification(hidden_layers_structure=[256, 256],
    learning_rate_rbm=0.05,
    learning_rate=0.1,
    n_epochs_rbm=10,
    n_iter_backprop=100,
    batch_size=32,
    activation_function='relu',
    optimization_algorithm='sgd',
    dropout_p=0.2)

# 如果只需调用训练好的模型，需要注释下面2行
# classifier.fit(train_data, train_label)
# Save the model
# classifier.save('supervised_model.pkl')

# Restore it
classifier = SupervisedDBNClassification.load('supervised_model.pkl')

I got error as title shows:

  File "e:\applications\wpy64-3741\python-3.7.4.amd64\lib\site-packages\dbn\models.py", line 183, in _compute_hidden_units_matrix
    np.dot(self.W, np.transpose(matrix_visible_units)) + self.c[:, np.newaxis]))

  File "<__array_function__ internals>", line 6, in dot

ValueError: shapes (256,33626) and (33227,800) not aligned: 33626 (dim 1) != 33227 (dim 0)

What's the problem?